arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2501.01425cs.CV
收藏

自由形式运动控制:在视频生成中控制相机与物体的6D位姿

Free-Form Motion Control: Controlling the 6D Poses of Camera and Objects in Video Generation

Fudan University(复旦大学);DAMO Academy, Alibaba group(阿里集团 DAMO 院);Hupan Lab(虎派实验室);Nanyang Technological University, Singapore(新加坡南洋理工大学)

Xincheng Shuai, Henghui Ding, Zhenyuan Qin, Hao Luo, Xingjun Ma, Dacheng Tao

AI总结:

本文针对视频生成中相机与物体6D位姿难以同时控制的问题,提出合成数据集SynFMC和自由形式运动控制方法FMC,实现三维感知的独立或联合位姿控制,并兼容多种个性化T2I模型,在多个场景中取得优于现有方法的效果。

摘要:

在生成视频中控制动态物体和相机的运动是一项有意义但具有挑战性的任务。由于缺乏带有全面6D位姿标注的数据集,现有的文本到视频方法无法以三维感知的方式同时控制相机和物体的运动,导致对生成内容的可控性有限。为了解决这一问题并推动该领域的研究,我们引入了一个用于自由形式运动控制的合成数据集(SynFMC)。所提出的SynFMC数据集包含多样的物体和环境类别,并根据特定规则覆盖各种运动模式,模拟常见和复杂的真实世界场景。完整的6D位姿信息有助于模型学习在视频中解耦物体与相机的运动效应。为了提供精确的三维感知运动控制,我们进一步提出了一种在SynFMC上训练的方法,即自由形式运动控制(FMC)。FMC能够独立或同时控制物体和相机的6D位姿,生成高保真视频。此外,它兼容各种个性化文本到图像(T2I)模型,以支持不同的内容风格。大量实验表明,所提出的FMC在多个场景中优于先前的方法。

英文摘要:

Controlling the movements of dynamic objects and the camera within generated videos is a meaningful yet challenging task. Due to the lack of datasets with comprehensive 6D pose annotations, existing text-to-video methods can not simultaneously control the motions of both camera and objects in 3D-aware manner, resulting in limited controllability over generated contents. To address this issue and facilitate the research in this field, we introduce a Synthetic Dataset for Free-Form Motion Control (SynFMC). The proposed SynFMC dataset includes diverse object and environment categories and covers various motion patterns according to specific rules, simulating common and complex real-world scenarios. The complete 6D pose information facilitates models learning to disentangle the motion effects from objects and the camera in a video.~To provide precise 3D-aware motion control, we further propose a method trained on SynFMC, Free-Form Motion Control (FMC). FMC can control the 6D poses of objects and camera independently or simultaneously, producing high-fidelity videos. Moreover, it is compatible with various personalized text-to-image (T2I) models for different content styles. Extensive experiments demonstrate that the proposed FMC outperforms previous methods across multiple scenarios.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑